United States - English U.S Change

Public lighting: switch it off or smarter lighting?

Public lighting: switch it off or smarter lighting?

For some years now, national and regional authorities have switched off public lighting on roads, either completely or more commonly, during parts of the night to reduce energy costs in public spending. Each time, concerns regarding safety have been raised and shortly afterwards, the street lights are switched on again.

At Schréder, we truly believe that the implementation of new technologies on all levels (i.e. luminaires, light sources and control systems), can generate energy savings of the same or an even higher amount compared to simply switching off, without putting the safety of drivers and passengers at risk.

Public lighting and road safety: review of major studies

The debate of lighting streets and highways has existed since it was decided to light them at night. The first recognised study was written in France, in 1935 [Geets R. (1980)].
It established even at that time that “the lighting of major road networks is a requirement”.

The Belgian Experience

Since 1968, a census to determine whether to light roads based on traffic volume criteria has been conducted in Belgium. The goal is to find the right method to decide whether to light or not with the emphasis on safety. In 1978, the first results, for the lit sections, were published [Geets R. (1980)].
Globally, a 30% drop in accidents was confirmed. This drop is more significant for injuries and deaths.

The same Geets report mentions that highway lighting represents less than 0.1% of the state’s budget, including investments and maintenance. The consumption represents only 0.07% of the total energy consumption of Belgium.

However in 1981, the Belgian government decided to not only switch off highway lighting for part of the night, but to also halve the luminance levels by disconnecting halve of the luminaires during the start-up period. A statistical study of the results of this decision on the rate of accidents was published in 1987 [De Clercq, G. (1987)].

This study confirmed the conclusions from the previous studies that street lighting has a positive influence on the number of accidents, serious injuries and deaths.

Public lighting switched off


 Serious injuries


 Public lighting reduced


 Serious injuries


Table 1: Percentage of additional accidents for the 2 implemented scenarios

The French Experience

The AFE (French Lighting Agency) requested the CNRS (National Centre for Scientific Research) to undertake a study between 1998 and 2002 [AFE (2009)]. This time, the study did not take into account any statistics on the ground but only drivers’ reactions in their vehicles. 

This experiment, done on a simulator, removed factors skewing the statistics whilst ensuring the safety of the person running the test in case of accident. Through these studies, they were able to verify whether street lighting contributed to improved safety, or not. For example, during the study, a potential accident situation was created to test the driver’s reactions.


Good anticipation


Not lit

4 drivers

4 drivers


10 drivers

2 drivers

Table 2: Results on the driver’s anticipation

The Norwegain Experience

One of the most important papers from the last few years is a PhD thesis written in Norway by P.O. Wanvik. Street lighting at night reduces accidents with injuries by 30%. Major effects are:

(1) 60% reduction of fatal injuries
(2) 45% reduction of accidents involving pedestrians resulting in injuries
(3) 50% reduction of accidents with injuries on highways.

These studies prove that street and highway lighting is required to improve users’ safety.
However, the decision to light or not, remains in the hands of public authorities.
Many tools exist today (norms, sources, technologies…) to enable lighting authorities to light the right place, with the right level at the right time! All this, whilst complying with the economic and safety specifications! 


Maintain lighting and generate savings: an example

While lighting and methods of lighting have evolved significantly since roads were first lit, the type of light sources has remained unchanged over 30 years: the low-pressure sodium lamp. Admittedly, its lighting efficiency (lumen/watt) remains unmatched. But today, it is not only the efficiency of the source that needs be considered.
Other types of sources have evolved in various aspects: light efficiency, color temperature, color rendition, lifetime and finally, the dimming ability versus lifetime.
Electronics makes it possible to offer new features (constant lumen output, adjustable lumen output…). 
Control options (unidirectional and bidirectional) can manage not only the light flow, but also the light in its totality (running time, source mortality, failures…).
Recent progress brings an optimized maintenance for public lighting.

This is the one domain where we can still improve: managing the lighting levels based on the needs at the time it is required.

Take for example, a road where the public lighting needs to be replaced. The public authorities are asking for several technical solutions for the project. These solutions must satisfy several needs:

  • Retrofit HPS (high-pressure sodium) or LED
  • Estimation of annual energy consumption 

In this context, we add two complementary criteria to estimate the efficiency of the luminaire or the installation:

  • Service level: percentage of users (vehicles, pedestrians, cyclists, …) who benefit from the light at night
  • Loss level: ratio between the accumulative time of absence of users when the light is switched on and the night duration.

For this simulation, we take into account several hypotheses:

  • Conformity with NBN 13201
  • Average night of 12h
  • Night traffic mostly concentrated over 4h 
  • 6 lighting level modulation scenarios


Night profile


The first 3 scenarios are set without specific equipment; it is a simple gradation or a switch off.

The other 3  scenarios need a detector to manage the switching on and off of the lights based on the presence or absence of an user. As the switching on and off phases are quick and unpredictable when a sensor is used, only the LED solution will be calculated for the last 3 scenarios.

Some principles were taken in the hypotheses for this comparison (for example average night duration), however they remain the same for all scenarios and types of sources.



100% throughout the night


50 % from 10pm to 6am


0% from 10pm to 6am


Detection from 10pm to 6am


Detection throughout the night


Detection throughout the night  + 50% from
10pm to 6am

Figure 1: Definition of scenarios for the comparison of solutions


The existing lighting for the area is provided by 7 luminaires fitted with 100W high-pressure sodium lamps and 22 fixtures equipped with 125W high-pressure mercury vapour lamps.
The recommended HPS solution is composed of 31 luminaires with 100W lamps.
The LED solution incorporates 38 fixtures of 54W.

Based on the imposed hypotheses, we can establish the annual consumption of the existing solution. We can also establish the two replacement solutions as well as solutions for each scenario.

Figure 2: Comparison of annual energy use per scenario

Public lighting is a service to the population. This service must be available 100% of the time.
However, we estimate that these luminaires do not need to be lit when there is no one present. It is at this moment that the service and loss levels come into play. We must maximize the former whilst minimizing the latter.
Figure 3 shows the ratio of these two criteria in the different scenarios.


Figure 3: Service and loss levels for the different scenarios

Figure 4: Financial losses for the two solutions based on the scenario

The estimation of financial losses [Figure 4], highlights losses generated by keeping lights on when there is no one around.
LEDs can help achieve zero financial loss, something that traditional sources cannot achieve as they do not allow for constant detection.
Again, the choice of management type is critical in this comparison.
For example, financial losses are the same if the lights are switched off from 10pm until 6am or for lighting equipped with detection over this period. However, the service level rendered to the population is not the same. 


Street and highway lighting is required to ensure safety for the general public and public services (police, ambulances...).
Today, it is possible to adapt public lighting to the needs of each user.
A light management system suitable for the needs of users can also achieve the financial savings required by public authorities.
Simply switching off existing equipment is not the solution in the long term.
Technological progress offers the possibility, in this period of economic crisis, to put in place structural savings in the public lighting sector.

For more details, please read the entire dossier.

Print this page

Your browser is currently not supported

Don't worry, there is an easy fix. All you have to do is click the link below and follow the instructions.